Today, few question the need for substantial improvement in infrastructure if India is to sustain its rapid growth. The Planning Commission has put front and centre in its strategy to sustain faster, broad-based and inclusive growth, an investment of about $500 billion in infrastructure before the end of the 11th Five Year Plan in March 2012.

Over the past three years, India has invested an average of 4.5 per cent of GDP each year in infrastructure. Getting to the Planning Commission's target - which is estimated to be the equivalent of 9 per cent of GDP in 2011-12 - means infrastructure investment will need to increase by nearly 1 per cent of GDP each year between 2007-08 and 20011-12. The question is: where will the resources to finance this additional investment�nearly 5 percentage points of GDP by 2012 - come from?

Let us look first at domestic sources. India's household sector saves about 24 per cent of GDP, up from around 16 per cent of GDP only a decade ago. Over this same period, India's corporate sector's savings has also increased from around 4 per cent of GDP to close to 8 per cent. Can we expect such increases to continue? Perhaps, but it is not likely. Household savings are already at the high end of broadly comparable emerging markets. And corporate profitability - which has been remarkably high - is likely to be buffeted by the slowing global economy, the financial turmoil, and upward pressure on wages. Therefore, it may not be realistic to expect much by way of additional savings from India's private sector.

What about the public sector? From dissavings of around 1 per cent of GDP a decade ago, public sector savings shot up to an impressive 3 per cent of GDP in 2007-08. But given the pressures from the Sixth Pay Commission, the agricultural loan waiver, and continued commodity subsidies, it will likely be difficult to increase or even to maintain this level of savings in the absence of substantial improvements in tax revenue collection.

So, the 500-billion dollar question: can India achieve the target for infrastructure investment without foreign capital? We argue that it cannot.

In 2006-07, net capital inflows amounted to about 5 per cent of GDP, but the residual gap between investment and domestic savings was only around 1 per cent of GDP. This led to the view amongst analysts and some policymakers that capital flows are "excessive" and should be curbed.

There is no denying that a major policy challenge is to maintain macroeconomic stability in the face of capital inflows that may not match outflows in timing. As investment is typically pre-funded, capital will flow in before it goes out as imports and profits. And this is true for FDI, portfolio inflows and fixed-income lending. Such asynchronised timing makes life very difficult for macroeconomic policymakers.

Often, and perhaps naturally, policymakers tend to give a much higher weight to short-term problems than to the longer view, and this leads to attempts to fine-tune capital controls to manage the impact of inflows. But fine-tuning capital controls rarely works. To quote Glenn Stevens, Governor of the Reserve Bank of Australia, as he discussed Australia's experience in the 1980s, "[Capital inflows] occurred in spite of capital controls that were still in place, because the distinction between current and capital transactions was blurring and market participants were becoming more adept at circumventing the controls."

Another rationale that is often provided for active capital account management is the belief that by changing relative prices (through taxes and other forms of transaction costs), policymakers can influence the form of the inflows, that is, foreign direct investment, portfolio investment or borrowings.

The choice of such taxes is derived from a hierarchy of the relative desirability of different types of inflows. There is little by way of evidence (even from the famed Chilean experiments with controls) that government policy can do better than the market in pricing different forms of capital.

At the same time, there is considerable evidence that controls do at least two things: they make capital inflows less transparent by incentivising agents to disguise them, and they raise the price of capital, especially for small and medium-term enterprises. This would appear to be at cross purposes with the need for large capital inflows to ensure the continuance of India's high growth.

Ultimately, the only way capital controls are known to have worked is when they were imposed in a draconian and blanket fashion. Fortunately, there is no appetite in India for such extreme measures.

Quo vadis, then? Given India's strong medium-term fundamentals, it is quite clear that foreign investor interest in India is here to stay. And the timing of such heightened interest is indeed fortuitous as it coincides with the government's recognition of the urgent need to kick infrastructure investment into high gear. So the right strategy is to rapidly put in place reforms of financial, labour and product markets to absorb these inflows efficiently.

India needs to rethink its capital account framework in the light of the need for $500 billion in infrastructure investment. This means refraining from ad hoc changes in capital controls in the name of macroeconomic management, and quickly expanding the country's real and financial absorptive capacity.

In particular, there is an urgent need for an expedited and time-bound plan to develop the corporate bond market, including by raising the limits on foreigners' participation in this market, permitting greater capital outflows, creating fiscal space to finance infrastructure investment by curbing wasteful spending, and implementing structural reforms of labour and product markets.

The author is a Senior Advisor in the Asia and Pacific Department of the International Monetary Fund. The views expressed here are personal